# RLHF: Research Digest

> 10 foundational papers with 10,643 combined citations, plus the latest arXiv preprints.

- **Canonical:** https://agentflare.org/scholar/rlhf-research-digest.html
- **Updated:** 2026-10-18
- **Category:** scholar
- **Full structured data:** `/scholar/rlhf-research-digest.data.json` — $0.02 via AISA HTTP 402 (https://esa.aisa.one/api/v1/access/verify; agents set crawlerAutoPrice=true)

## Key data

- **Field:** reinforcement learning from human feedback
- **Foundational:** 10
- **Recent (arXiv):** 8
- **Updated:** 2026-10-18

10 foundational papers with 10,643 combined citations, plus the latest arXiv preprints.

_…full analysis and the complete dataset are available to agents for $0.02 — fetch `/scholar/rlhf-research-digest.data.json` (HTTP 402)._

## Sources

1. [Open problems and fundamental limitations of reinforcement learning from human feedback](https://arxiv.org/abs/2307.15217)
2. [Training a helpful and harmless assistant with reinforcement learning from human feedback](https://arxiv.org/abs/2204.05862)
3. [Rlhf deciphered: A critical analysis of reinforcement learning from human feedback for llms](https://dl.acm.org/doi/abs/10.1145/3743127)
4. [A survey of reinforcement learning from human feedback](https://arxiv.org/abs/2312.14925)
5. [Principled reinforcement learning with human feedback from pairwise or k-wise comparisons](http://proceedings.mlr.press/v202/zhu23f.html)
6. [Rlaif vs. rlhf: Scaling reinforcement learning from human feedback with ai feedback](https://arxiv.org/abs/2309.00267)
7. [Rlaif: Scaling reinforcement learning from human feedback with ai feedback](https://openreview.net/forum?id=AAxIs3D2ZZ&utm_source=chatgpt.com)
8. [Reinforcement learning from human feedback](https://arxiv.org/abs/2504.12501)
9. [Safe rlhf: Safe reinforcement learning from human feedback](https://proceedings.iclr.cc/paper_files/paper/2024/hash/dd1577afd396928ed64216f3f1fd5556-Abstract-Conference.html)
10. [Personalizing reinforcement learning from human feedback with variational preference learning](https://proceedings.neurips.cc/paper_files/paper/2024/hash/5e1c255653eb98cef13f45b2d337c882-Abstract-Conference.html)
11. [Metric-Driven Reinforcement Learning for Object Hallucination Reduction in LVLM-Based Image Captioning](https://arxiv.org/abs/2610.14976v1)
12. [A Scoping Review and Experimental Study on Reinforcement Learning from Human Feedback for Human-Robot Collaboration](https://arxiv.org/abs/2610.09891v1)
13. [AI Safety via Debate is Compromised by Cognitive Biases](https://arxiv.org/abs/2610.05461v1)
14. [Groupwise Distortion Guarantees for Preference-Based Alignment](https://arxiv.org/abs/2610.05450v1)
15. [Beyond Refusal Patterns: Safe-Role Internalization for Robust and Generalizable LLM Safety Alignment](https://arxiv.org/abs/2610.07023v1)
16. [Using Context Is Not Enough: Test-Time Training for Personalized Reward Modeling](https://arxiv.org/abs/2609.35109v1)
17. [RMB: Reward Model Boosting Mitigates Reward Hacking](https://arxiv.org/abs/2609.33221v1)
18. [Distortion of AI Alignment Revisited: RLHF is a Decent Utilitarian Aligner](https://arxiv.org/abs/2609.12651v1)

## Related

- [LLM Agents & Planning: Literature Digest](https://agentflare.org/scholar/llm-agents-planning-literature-digest.html)
- [Retrieval-Augmented Generation: Research Digest](https://agentflare.org/scholar/retrieval-augmented-generation-research-digest.html)
- [AI Alignment & Safety: Research Digest](https://agentflare.org/scholar/ai-alignment-safety-research-digest.html)
- [Multimodal Foundation Models: Research Digest](https://agentflare.org/scholar/multimodal-foundation-models-research-digest.html)
- [Mechanistic Interpretability: Research Digest](https://agentflare.org/scholar/mechanistic-interpretability-research-digest.html)
- [Tool Use in Language Models: Research Digest](https://agentflare.org/scholar/tool-use-in-language-models-research-digest.html)

---
_Part of AgentFlare, an agent-native data network powered by AISA. https://aisa.one/docs_